The Promise and Pitfalls of Conflict Prediction: Evidence from Colombia and Indonesia

A-Tier
Journal: Review of Economics and Statistics
Year: 2022
Volume: 104
Issue: 4
Pages: 764-779

Authors (6)

Samuel Bazzi (Boston University) Robert A. Blair (not in RePEc) Christopher Blattman (not in RePEc) Oeindrila Dube (not in RePEc) Matthew Gudgeon (Tufts University) Richard Peck (not in RePEc)

Score contribution per author:

0.670 = (α=2.01 / 6 authors) × 2.0x A-tier

α: calibrated so average coauthorship-adjusted count equals average raw count

Abstract

How feasible is violence early-warning prediction? Colombia and Indonesia have unusually fine-grained data. We assemble two decades of local violent events alongside hundreds of annual risk factors. We attempt to predict violence one year ahead with a range of machine learning techniques. Our models reliably identify persistent, high-violence hot spots. Violence is not simply autoregressive, as detailed histories of disaggregated violence perform best, but socioeconomic data substitute well for these histories. Even with unusually rich data, however, our models poorly predict new outbreaks or escalations of violence. These "best-case" scenarios with annual data fall short of workable early-warning systems.

Technical Details

RePEc Handle
repec:tpr:restat:v:104:y:2022:i:4:p:764-779
Journal Field
General
Author Count
6
Added to Database
2026-01-24